Affective disorder appears to be strongly influenced by genetic factors. However, apart from recent DNA marker studies in certain families, particular biological markers of vulnerability have yet to be confirmed. Suppression of melatonin by light has been found to be unusually sensitive in a small number of medications. It appears to be similarly sensitive in some high-risk offspring of bipolars. Thus, it may be a trait marker for certain forms of affective disorder. This work requires replication in a new population. Using a case-control design, we propose to test melatonin suppression by light in euthymic bipolar patients, examine its relationship to baseline melatonin rhythms on a dark night in the same subjects, and compare light suppression in patients to that in controls for sex, age, and season of testing. Melatonin suppression will be measured by exposing subjects to 500 lux light between 2 A.M. and 4 A.M. and measuring plasma melatonin before and after the light stimulus. The study will be performed on a clinical research ward. We also propose to test reproducibility of the melatonin suppression response in volunteers and examine the effects of lithium on suppression in patients. A group of unipolar patients would also be tested to assess the response in another affective subgroup. These patients will also be compared with matched controls. Confirmation of a biologic vulnerability marker for bipolar illness and related affective conditions would be an important step in understanding their pathophysiology. It would also be potentially useful in genetic counselling for patients and their families.